EP0422544B1 - Appareil pour le traitement de données d'images en couleurs - Google Patents

Appareil pour le traitement de données d'images en couleurs Download PDF

Info

Publication number
EP0422544B1
EP0422544B1 EP90119238A EP90119238A EP0422544B1 EP 0422544 B1 EP0422544 B1 EP 0422544B1 EP 90119238 A EP90119238 A EP 90119238A EP 90119238 A EP90119238 A EP 90119238A EP 0422544 B1 EP0422544 B1 EP 0422544B1
Authority
EP
European Patent Office
Prior art keywords
hue
color
value
picture element
chromatic color
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Lifetime
Application number
EP90119238A
Other languages
German (de)
English (en)
Other versions
EP0422544A2 (fr
EP0422544A3 (en
Inventor
Yasuhiko Nakano
Shigeru Yoshida
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Fujitsu Ltd
Original Assignee
Fujitsu Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Fujitsu Ltd filed Critical Fujitsu Ltd
Publication of EP0422544A2 publication Critical patent/EP0422544A2/fr
Publication of EP0422544A3 publication Critical patent/EP0422544A3/en
Application granted granted Critical
Publication of EP0422544B1 publication Critical patent/EP0422544B1/fr
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—2D [Two Dimensional] image generation
    • G06T11/001—Texturing; Colouring; Generation of texture or colour
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/46—Colour picture communication systems
    • H04N1/56—Processing of colour picture signals
    • H04N1/60—Colour correction or control
    • H04N1/62—Retouching, i.e. modification of isolated colours only or in isolated picture areas only

Definitions

  • the present invention relates to a color image data processing apparatus for processing color image data, and in particular, one that can adjust the color blur of a chromatic color at a time of reading processing color image data to accurately distinguishing a character used on a color document such as an OCR (optical character recognition) slip from a colored form.
  • OCR optical character recognition
  • a format such as that of a character frame of an OCR for reading hand written characters is printed in a special color called a dropout color. This format is optically eliminated at a reading time so that only the character is recognized.
  • the character frame of an OCR slip that sets the position of the written character is printed in a dropout color that can be clearly identified by the human eye, but cannot be distinguished from the background color by an OCR sensor.
  • the optical filter determined by the spectral characteristics of the written character color and those of the dropout color is set in front of the OCR sensor. The character frame is thereby optically eliminated when the original is read, so that only the character to be recognized is read.
  • such an existing OCR has the problem of having to select the characteristics of the dropout color and the optical filter according to the means of writing the character. Furthermore, it completely fails to use the information indigenous to the character frame, since the character frame printed in the dropout color is eliminated from the image data at a time of reading. For example, if the character frame is also recognized, the cutout in a character unit necessary for a character identification is quite easily executed. Yet, the existing OCR has the problem that it cannot use the character frame information at all. This problem becomes extremely large when an OCR slip is printed in multiple dropout colors so that it contains further effective information.
  • the method in this proposal is an HSV type identification method, in which an original such as a slip is read as an RGB (red, green and blue) image signal that is converted to an HSV (hue, saturation, value) signal. After an achromatic color and a chromatic color are separated by a predetermined saturation threshold, the value of the achromatic color is used for a character identification and the chromatic color is outputted as a two value image, or a multiple value image corresponding to the number of hues separated by a plurality of predetermined thresholds.
  • RGB red, green and blue
  • HSV hue, saturation, value
  • a background, a character, and a character frame are indicated. That is, as shown in Figure 1A, since the background and the character are each of an achromatic color, they come into an inner solid cylindrical region of the HSV coordinate system. Furthermore, since the character frame is of a chromatic color, it comes into an outer hollow cylindrical region. After the HSV conversion, these characteristics are used first to separate the background and the character, each being of an achromatic color, from the character frame of a chromatic color by a saturation threshold. Then the background and the character are separated from the separated achromatic color by a value threshold. Meanwhile, the chromatic color is identified as a multiple value image of a predetermined color number by the predetermined thresholds for separating the hue of the representative color used on a slip, as shown in Figure 1B.
  • a image reading part IR for reading an original OR such as a colored slip and a color identification part CI are conventionally set as shown in Figure 2.
  • the image reading part IR comprises lenses L1, L2 and L3; a red color filter RF for passing a red light through; a green color filter GF for passing a green light through; a blue color filter BF for passing a blue light through; a photoelectric element of a picture element unit, such as charge coupled devices (CCDs) C1, C2 and C3 for detecting the red, green and blue light, respectively; amplifiers A1, A2 and A3; and A/D converters ADC1, ADC2 and ADC3 for converting an analog signal to a digital signal.
  • the color identification part CI comprises an HSV conversion circuit CONV; comparison circuits COM1, COM2 and COM3 and multiplexers MUX1 and MUX2.
  • the lenses L1 , L2 and L3; the red color filter RF; the green color filter GF; the blue color filter BF; the CCDs C1 , C2 and C3; the amplifiers A1, A2 and A3; and the A/D converters ADC1, ADC2 and ADC3 of the image reading part IR configure three optical signal systems each for a red signal, a green signal and a blue signal.
  • These signals read the original OR in an image signal line unit worth of three colors, convert the image signal in the line to a digital signal in one picture element unit, and transmit the red, green and blue signals to the HSV conversion circuit CONV via signal lines R1, G1 and B1.
  • the HSV conversion circuit CONV converts the red, green and blue signals to S, V and H signals and outputs them.
  • the S signal is inputted to a conversion circuit COM1 and compared with a predetermined threshold T S .
  • this threshold T S is for separating an achromatic color from a chromatic color. If the S signal is smaller than this threshold T S , it is recognized as a background or a character; if the S signal is equal to or greater than this threshold T S , it is recognized as a character frame.
  • the comparison circuit COM1 outputs e.g. "0" for an achromatic color if S ⁇ T S and "1" for a chromatic color if S ⁇ T S .
  • the V signal is inputted to a conversion circuit COM2 and compared with a predetermined threshold T V .
  • This threshold T V is for distinguishing the value of the background from that of the character of the achromatic color. If the V signal is smaller than this threshold T V , it is recognized as a black part, i.e. a character part; if the V signal is equal to or greater than this threshold T V , it is recognized as the background.
  • the comparison circuit COM2 outputs e.g. "1" for a character if V ⁇ T V and "0" for a background if V ⁇ T V .
  • the achromatic color part of the output signal from this comparison circuit COM2 constitutes the background and the character.
  • This part is selected by multiplexer MUX1 controlled by the output from the comparison circuit COM1 and is outputted “as is” if S ⁇ T S .
  • the chromatic color part constitutes the character frame.
  • This part is outputted as "0”, transmitted to the multiplexer MUX1 and eliminated. This is similar to it being read by using a dropout color if S ⁇ T S .
  • An output signal S4 from the multiplexer MUX1 is transmitted to a character recognition part of its latter stage (not shown in the drawing).
  • the H signal is inputted to a conversion circuit COM3 and compared with a predetermined series of thresholds T H .
  • this series of thresholds T H comprises T1 through T5, which distinguish red from yellow, yellow from green, green from cyan, cyan from blue, and blue from magenta, respectively. If the H signal is smaller than threshold T1, it is recognized as red; if the H signal is equal to or greater than this threshold T5, it is recognized as magenta.
  • the character frame is of one color, namely, if a format is of one color, no distinction has to be made. However, if it is of two colors, e.g. when two kinds of forms need to be distinguished, one threshold is required. If it is of three colors, two thresholds are required. Similarly, if six colors need to be distinguished, five thresholds are required. The number and value of the threshold(s) depends on the number of colors used in the slip.
  • multiplexer MUX2 outputs "as is” the outputs i.e., character frame information, from the comparison circuit COM3, when "1" is transmitted from the comparison circuit COM1, and outputs "0" to multiplexer MUX2 when "0" is transmitted from the comparison circuit COM1, i.e., S ⁇ T S , in case of an achromatic color.
  • An object of the present invention is to offer a new color image data processing apparatus for eliminating a chromatic color change arising at a chromatic color edge part in color image data.
  • a feature of the present invention resides in a color image data processing apparatus means for processing hue information of each picture element read out sequentially comprising a chromatic color detection means for detecting a chromatic color picture element series that is a consecutive succession of chromatic color picture elements in a main scanning line, and a replacing means for replacing the hue value of each picture element of the chromatic color picture elements series with the hue value of a picture element located approximately at a center position of the detected chromatic color picture element series.
  • Figure 3 is a configuration diagram of the first principle of this invention.
  • FIG 3 1 is a color image data processing device equipped with this invention, comprising a color image data reading portion 2, a color information computing portion 3, a chromatic color picture element series detecting portion 4, a hue value replacing portion 5, and a hue value averaging portion 6.
  • 7 is an original on which multiple color image data are written.
  • the color image data reading portion 2 optically reads out in a sequence color information of the RGB color specification system that each picture element of color image data of the original 7 owns.
  • the color information computing portion 3 computes color information of an HSV color specification system corresponding to the color information of each picture element that the color image data reading portion 2 reads out, and outputs a corresponding hue value to a picture element of a chromatic color and a zero hue value to a picture element of an achromatic color.
  • the hue value outputted from the color information computing portion 3 is inputted to the chromatic color picture element series detecting portion 4 that detects a chromatic color picture element series, i.e., a succession of series of chromatic color picture elements, at a main scanning line.
  • the chromatic color picture element series detected by the chromatic color picture element series detecting portion 4 is inputted to the hue value replacing portion 5, which finds the hue value of the picture element positioned approximately at the center of the detected chromatic color picture element series.
  • the hue value replacing portion 5 then replaces the hue value of the detected chromatic color picture element series with this found hue value, as much as possible.
  • the hue value outputted from the hue value replacing portion 5 is inputted to the hue value averaging portion 6, which finds the average of the hue values of each picture element and a picture element positioned in a region near the each picture element, replaces the hue value of the each picture element with this found average value of the hue value and outputs them.
  • the chromatic color picture element series detecting portion 4 detects a series of chromatic color picture elements, i.e., a succession of picture element that has the same color, at the main scanning line, by detecting a change in the hue value of the picture element being inputted from zero to a value other than zero, and a change in the hue value of the picture element from a value other than zero to zero.
  • the picture element or a plurality thereof corresponding to an edge part of the chromatic color picture element series detected at this time often has a hue value that is different from the original hue value.
  • the hue value replacing portion 5 adjusts the color blur of a chromatic color arising in the main scanning line direction, by finding the hue value of the picture element positioned approximately at the center of the detected chromatic color picture element series and replacing the hue value of each picture element of the detected chromatic color picture element series with the hue value of this center location.
  • the hue value averaging portion 6 adjusts the color blur of chromatic color occurring in the sub-scanning line direction by finding the average of the hue values of the targeted picture element that must be processed and the picture element positioned in a region near the targeted picture element by using the hue value of each of these replaced picture elements and replacing the hue values of the targeted picture element with the found average of the hue value.
  • the first embodiment of the invention eliminates a chromatic color change arising in a chromatic color edge part in the color image data.
  • the first embodiment of the invention is explained in detail referring to the embodiment of an application to an OCR system.
  • Figure 4 shows the configuration of the hue-value reading processing part of the color image data processing device 1 related to an embodiment of this invention.
  • Figure 5 shows the configuration of the processing part, in the color image processing device 1, for adjusting the color blur contained in the hue value outputted from the embodiment configuration of Figure 4.
  • 10 is an image reading device part that optically reads color image data comprising character 7a (written in an achromatic color of a low value) and character frame 7b (printed in multiple chromatic colors) on the original which is transmitted at a constant speed by transmitting apparatus not shown.
  • the image reading device part 10 sequentially reads the RGB color information of the original 7's color image data in a line unit according to three optical systems comprising three lenses 11a, 11b and 11c for an image formation; three optical filters 12a, 12b and 12c for red, green and blue, conforming to an RGB color specification system; and three line image sensors 13a, 13b and 13c for reading image data. It then converts them to digital signals with A/D converters 15a, 15b and 15c after amplifying them with amplifiers 14a, 14b and 14c, and transmits the RGB signal value of each picture element to a color information conversion device part 20.
  • the color information conversion device part 20 finds the information on a hue, a saturation and a value of each picture element of image data read by the image reading device part 10, and separates the character 7a of the original 7 (in an achromatic color of a high value) and the character frame 7b of the original 7 from the background. After receiving the RGB signal value of each picture element from the image reading device part 10, the color information conversion device part 20 finds the hue value H, the saturation value S and the value V of each picture element according to the HSV conversion circuit 21.
  • the color information conversion device part 20 judges whether the color information of the picture element is of an achromatic color or a chromatic color by comparing this found saturation value S with the predetermined threshold T S at a saturation comparison circuit 22, and outputs "0" for an achromatic color if S ⁇ T S and "1" for a chromatic color if S ⁇ T S .
  • the color information conversion device part 20 judges whether the color information of the picture element is of a background or a character by comparing this found value V with the predetermined threshold T V at a value comparison circuit 23, and outputs "1" for a character if V ⁇ T V and "0" for a background if V ⁇ T V .
  • An "AND” gate 24 makes a logical product by multiplying the inversion of the output from the saturation comparison circuit 22 outputted from an inverter 25 and the output from the value comparison circuit 23, and outputs "1" for a character 7a and "0" for a background or a character frame 7b, so that a cutout of the character 7a is processed.
  • the hue value H outputted from the HSV conversion circuit 21 of the color information conversion device part 20 is sometimes of a different chromatic color from the original chromatic color at an image edge part, as described earlier.
  • the embodiment of the processing part for adjusting the hue value of the color image data reading device 1 related to the first invention shown in Figure 5 shows how the color blur of this chromatic color is adjusted. The process of adjusting the color blur of the chromatic color that this invention executes according to the embodiment shown in Figure 5, is explained.
  • a gate circuit 30 After receiving a code for identifying a chromatic color and an achromatic color and the hue value of each picture element read out in a unit of the main scanning line direction from the color information conversion device part 20, a gate circuit 30 executes a process of storing the hue value of each picture element of one main scanning line in a line buffer 31 by outputting the hue value in case of a chromatic color and a zero hue value in case of an achromatic color. After this storing process is completed, a controller 33 reads out the hue values of each picture element from the line buffer 31 one by one and inputs them to a comparator 32. The comparator 32 detects whether these inputted hue values are zero or changing from zero to a value other than zero, or vice versa, and notifies the controller 33 of the detected result.
  • the controller 33 Upon receipt of a notification that the hue value being inputted from the comparator 32 is zero, the controller 33 stores this value in the memory part at the lowest stage of a five line memory 34 (for storing the hue value of each picture element of five main scanning lines). Meanwhile, upon receipt of a notification that the hue value of the picture element being inputted is detected to have changed to a value other than zero, by transferring the control to a chromatic color replacer 35, the controller 33 replaces the hue value of the subsequent chromatic color picture element (being the chromatic color picture element before the receipt of a notification from the comparator 32 that the hue value has changed to zero and corresponding to the series of chromatic color picture elements explained by Figure 3) with the hue value without a color blur.
  • a counter 36 of the chromatic color replacer 35 counts the picture element number of the chromatic color picture element series that must be processed on receipt of the comparison result of the comparator 32 from the controller 33, and a median setting part 37 of the chromatic color replacer 35 finds the hue value of the picture element positioned approximately at the center of the chromatic color picture element series that must be processed by selecting the hue value read out from the line buffer 31 according to the count value of this counter 36.
  • the median setting part 37 finds the hue value of the third picture element of the chromatic color picture element series, for which the median setting part 37 comprises e.g. multiple buffers and finds the hue value of the picture element positioned approximately at the center of the chromatic color picture element series by storing a candidate of the hue value of the center position of the chromatic color picture element series in those buffers in accordance with the notching of the count value of the counter 36.
  • the median setting part 37 finds the hue value of the center position of the chromatic color picture element series because the counter 36 finishes counting, a replacing part 38 of the chromatic color replacer 35 outputs the found hue value of the center position of the count value of the counter 36.
  • the controller 33 stores this replaced hue value in the memory part of the lowest stage of the five line memory 34.
  • Figure 6 explains the adjusting process that the chromatic color replacer 35 executes.
  • a chromatic color picture element series of "64, 70, 70, 70, 62 ⁇ with at least one "0" before and after the series is stored in the line buffer 31, the counter 36 counts "5".
  • the median "70” of the chromatic color picture element series is read out and stored five consecutive times as the replacing hue value in the line memory 34 via the controller 33.
  • the hue value after the adjustment becomes "70, 70, 70, 70, 70" with at least one "0" before and after the series.
  • the chromatic color replacer 35 realizes an adjustment of the color blur of the chromatic color arising in the main scanning line direction by replacing the hue value of the picture element at an edge part of the chromatic color picture element series to the hue value of the median location of the chromatic color picture element series.
  • the five line memory 34 stores the hue value of the newest main scanning line just read and the hue values of the four consecutive main scanning lines already read, according to the controller 33's actions.
  • an adder 39 After storing the hue values in the memory part of the lowest stage of the five line memory 34, namely, in the memory part for the newest main scanning line , an adder 39 adds the hue values of the five line memory 34 in the picture element unit of the sub-scanning direction by following the directive from the controlling mechanism 50, and stores the added result in a line memory 40.
  • a comparator 41 judges whether the hue value sequentially read out from the five line memory 34 is zero or is a value other than zero.
  • a counter 42 to which the judging result of the comparator 41 is inputted counts the number of chromatic color picture elements positioned near the targeted picture element set in the five line memory 34, according to a reset signal and an enable signal from the control mechanism 50. That is, when a 15 x 5 window (whose "5" is the line number of the five line memory 34) is, centered on the targeted picture element as shown in Figure 7, the number of the chromatic color picture elements in the window is counted.
  • the control mechanism 50 sets the targeted picture element to be processed sequentially, e.g. from the left of the center (the third line) memory part of the five line memory 34. Also, when the targeted picture element is close to the left end or the right end, the window form is different from 15 x 5.
  • an adder 43 computes the grand total of the hue value of the chromatic color picture elements positioned in the region near the targeted picture element (which is the same as the nearby region counted by the counter 42). By dividing this computed value of the adder 43 by the counted value of the counter 42, an averager 44 calculates and outputs the hue value averaged in the region near the targeted picture element. By repeating a similar process, the averaged hue value for all targeted picture elements is calculated. This averaged hue value absorbs a loss of color balance based on the difference of a picture element or a few picture elements in the sub-scanning line direction.
  • the control mechanism 50 executes the averaging of the hue value of the targeted picture element of the next main scanning line.
  • the averager 44 realizes an adjustment of the chromatic color blur arising in the sub-scanning line direction.
  • this invention is configured to adjust the chromatic color blur arising in the main scanning line direction according to the processings of the chromatic color replacer 35, as well as to adjust the chromatic color blur arising in the sub-scanning line direction according to the processings of the averager 44.
  • the present invention should in no way be construed as being limited to the preferred embodiments shown in attached drawings.
  • the method of replacing the hue value of the chromatic color picture element series and the method of averaging the hue values of the region near the targeted picture elements, for example, are realizable by various processing methods.
  • Figure 8 is a system flow chart of action processings that are executed when a processor replaces the hue value of the chromatic color picture element series and averages the hue values of the region near the targeted picture element.
  • the processor executes a gate processing step S1, and sets the hue value of the picture element at either "0" in case of an achromatic color or the inputted hue value in case of a chromatic color.
  • the processor judges whether the input is of a chromatic color picture element series (S2). If it is (YES), the processor executes the replacement (S3). In judging S2, if the input is not judged to be of chromatic color, or after replacing S3, the processor outputs a hue value H (S4) and writes the hue value in a line buffer (S5).
  • the hue value of the center picture element of a consecutive series thereof is set as the hue value of the series.
  • the processor deems the input to be of a chromatic color picture element series, in replacing S3 as shown in Figure 8, the third hue value from the beginning of the consecutive picture element series, for example, is set as the hue value of the series.
  • This hue value H is output at S4.
  • the processor judges whether or not one line of picture element data is gathered (S6). If it is not (NO), the processor repeats the steps from S1; if it is, i.e., it is inputted (YES), the processor moves the data in the line buffer to the line memory (S7).
  • Steps S2 through S6 described earlier are the executions of the processings of one picture element unit in the line direction.
  • the processor moves the hue value obtained in the moving S7 and judges whether the five lines of data are accumulated in the line memory (S8).
  • the processing of one line in the center is processed in a five-line unit.
  • the first line of data is stored in the line memory.
  • the second, third, fourth and fifth lines of data are stored in the line memory. Consequently, before the fifth line of data is stored, the five lines of data are judged not to have been accumulated in the judging S8 (NO), and the processings are repeated from S1.
  • the processor adds the hue values (H) at the same dot position in the five lines (S9), namely, adds five hue values in a vertical direction, to determine the hue value of the picture element on the center line.
  • This adding process S9 corresponds to the adding action of the adder 39.
  • the processor adds the hue values of the picture elements in a region near the targeted picture element, e.g.
  • Steps S10 and S11 correspond to the actions of the adder 43 and the counter 42 in Figure 5.
  • step S12 the processor moves its window to the other targeted picture element and its surrounding region (S13). If movement to the next targeted picture element S13 is possible, the processor judges that a subsequent line of data has not been processed (NO) in the judging step (S14) and repeats the processings from step S10, to process the next targeted picture element. When one more line of data has been processed, the processor judges whether or not all the picture elements have been processed. If they have not (NO), a subsequent line still exists, so the processor again repeats the processings from step S1. If they have all been processed (YES), the processor terminates all further processings.
  • Figures 9A and 9B show the principle of the second embodiment of the invention.
  • the second invention divides an image region P into windows W00 through W45, e.g., comprising sixteen (16) picture elements in a row and four (4) picture elements in a column for averaging the hue value of the chromatic color part in each window.
  • Figure 9A shows the condition in which 96 x 20 picture elements are divided into 6 x 5 thirty (30) windows.
  • the second embodiment of the invention is configured with an HSV conversion circuit 110 for converting RGB signals into HSV signals, a comparator 111 for comparing the S signal with a threshold T S for separating it into a chromatic color and an achromatic color and a hue value averaging part 112 for averaging the chromatic color hue value within a window.
  • This hue value averaging part 112 comprises an "AND" gate 113, an adder 114, a line memory 115, a register 116, a memory counter 117, a memory control part 118, and an averaging part 119.
  • the image region P shown in Figure 9A is scanned in the direction from the window W00 to W05, i.e. in the main scanning direction. Since a window is scanned by four lines at this time, windows W00 through W45 are scanned by scanning lines l0 through l3.
  • the RGB signals obtained from the scanning by scanning line l0 is input to the HSV conversion circuit 110 shown in Figure 9B and converted into HSV signals by each picture element.
  • the S signal is inputted to the comparator 111, which outputs "1" only for the input of the picture element part that is larger than the threshold T S , i.e., the picture element is judged to be of a chromatic color. Therefore, the "AND" gate 111 is turned ON, and the hue value H of this chromatic color is inputted to the adder 114.
  • One input of the adder 114 is "0" for the first line l0.
  • the line memory 115 Since the line memory 115 has a memory capacity for one line of picture elements in the image region P, the hue value H of the chromatic color corresponding to the picture elements of windows W00 through W05 is stored in the line memory 115.
  • the memory counter 117 counts the number of picture elements of the output "1" from the comparator 111.
  • the memory control part 118 reads and keeps the count value of the memory counter 117 by each of the 16 picture elements of the line l0.
  • the adding processing with the line l0 is performed by scanning in the line l1 direction in this manner. Then, by processing in the line l2 direction, the hue value H of picture element E20 is transmitted to the adder 114 through the "AND" gate 113 that adds the current hue value H of picture elements E00 and E01.
  • the line memory 115 keeps these four added results according to the scanning position correspondence of the picture elements. Then, the memory control part 118 keeps the data corresponding to the number of chromatic color picture elements of each block for scanning lines l0 through l3.
  • the memory control part 118 outputs to the averaging part 119 the hue value H kept in the line memory 115 corresponding to each window and the data depending on the number of chromatic color picture elements. As a result, the averaging part 119 computes the average of the hue values H of each window.
  • the average hue value of each window is calculated by the second embodiment of the invention, and the chromatic color part within the same window is replaced with this average value, thus averaging the hue value.
  • a color of a high hue value existing discretely at a certain edge part is diluted and the color distribution is narrowed. Hence, the color separation with a threshold becomes easier.
  • FIGS 10, 11A, 11B, 11C and 11D explain a preferred embodiment of this second invention.
  • Figure 10 shows a configuration of the preferred embodiment of this invention.
  • Figure 11A shows the details of the memory control part 118 and the averaging part 119.
  • Figure 11B shows the details of the table.
  • Figure 11C shows an action flow chart of the memory control part 118.
  • Figure 11D explains the memory access.
  • 101 is an image-reading part corresponding to the image reading part IR in Figure 2.
  • 102 is an original such as a colored slip, corresponding to the original OR in Figure 7.
  • 120 is a comparator for comparing the V signal that indicates the value with a threshold T V , and outputs "1" if the V signal is lower than the threshold T V .
  • 121 is an "AND” gate that outputs "1” because an inverter 122 outputs "1" when a comparator 120 outputs "1” and the comparator 111 outputs "0" by detecting an achromatic color. In other words, this "AND” gate 121 outputs a character signal "1".
  • 123 is a histogram memory for creating a hue histogram.
  • the hue values H averaged by the averaging part 119 of the window size (16 x 4 in the example shown in Figure 9A) are outputted to the histogram memory 123.
  • 124 is a threshold setting part that sets a threshold in the valley part of the hue histogram and creates a lookup table 125, based upon the histogram kept in the histogram memory 123.
  • the hue value H passed through the averaging part 119 refers to this lookup table 125, from the second page of the original, the color (color number) is determined.
  • this invention calculates the hue value average of each chromatic color part, for which it is necessary to find both the number of the picture elements of the chromatic color part in each window and the sum of the hue values H of this picture element part.
  • the memory control part 118 comprises a first counter 131 for counting the number of picture elements in a row of a window, which is four (4: 0 through 3) in this case; a second counter 132 for counting the number of picture elements in a column of a window, which is sixteen (16: 0 through 15) in this case; a table 133; and a control part 134.
  • the first counter 131 and the second counter 132 synchronize their counting actions with the scanning of lines l0 through l3.
  • the memory counter 117 to which "1" is outputted for a chromatic color picture element from the comparator 111 counts the number of chromatic color picture elements.
  • the first counter 131 counts zero (0).
  • the second counter 132 is invoked from zero, and when it has counted 15, the control part 134 reads the value of the memory counter 117, writes the counted value in the table 133 and resets the memory counter 117. Accordingly, as shown in Figure 11B, based on the scanning of the scanning line l0 for the windows W00, W01, etc.
  • the control, part 134 sequentially reads out the data A0 through A3 in the count compartment C0 for scanning lines l0 through l3 from the table 133 and outputs them sequentially to a second operation part 136 of the averaging part 119. Then, the second operation part 136 calculates the sum of A0 through A3 and temporarily keeps it.
  • the memory control part 118 sequentially outputs the data of the sixteen (16) picture elements corresponding to the window W00 from the line memory 115 through the register 116 to a first operation part 135, which is performed under the control of the control part 134 of the memory control part 118.
  • a sum of the hue values H at the window W00 is obtained.
  • the memory control part 118 sets an address value AD to zero (0) (S31), and reads the sum of the hue values H stored in the address AD of the line memory 115 (S32). This read value is stored in the register 116 and inputted to the adder 114.
  • the adder 114 adds the hue value H of a line's head end and the sum (total) stored in the register 116 (S33) and stores the result in the position of the read address AD (S34).
  • the hue value H of the targeted picture element of a new line is added to the same address of the line memory (a position on a line of the targeted picture element).
  • the memory control part 118 increments the address AD (S35), and it is judged whether the read address AD is greater than a final value END AD (S36). If it is not, steps from the reading S32 are repeated. After all the other steps, the judging step S36 terminates the addition to a line, if the address AD becomes greater than the final value END AD.
  • a window comprises 4 lines x 16 picture elements
  • an adding operation for one window is performed.
  • the sum of this hue values H is outputted to the second operation part 136 and then divided by (A0 + A1 + A2 + A3) to obtain the average of the hue values H of the chromatic color picture elements over the threshold T S .
  • control part 134 outputs the hue values H of the sixteen (16) picture elements corresponding to the window W01 from the data memory 115 of the count compartment C1 by the table 133 to the first operation part 135, and adds them to obtain the sum of the hue values H of the chromatic color picture elements in the window W01. This sum is outputted to the second operation part 136, where it is divided by (B0 + B1 + B2 + B3) to obtain the average of the hue values H of the chromatic color picture elements over the threshold T S .
  • the scan data of the next line's windows W10, W11, W12, etc. in the scanning line l0 through l3 are outputted sequentially, to which similar processings are performed to obtain the average of the hue values H for each window.
  • the hue histogram is created in the histogram memory 123.
  • the threshold setting part 124 sets a threshold in the valley part of the hue histogram, based on the histogram memory 123 for creating the lookup table 125. After the second original passes through the averaging part 119, the color is determined by referring to the lookup table 125. Since the histogram creation and the lookup table are not closely related to the objective of the second invention, their detailed explanation is omitted.
  • the chromatic color picture element number computing means in a window is not limited to the means shown in Figures 10, 11A, 11B, 11C and 11D, but instead can be a means such as that shown in Figure 12.
  • the memory counter 137 counts the number of chromatic color picture elements in a window for each scanning line. Then the memory control part 138 outputs "1" for every sixteen (16) picture elements to the "AND” gate 139 and turns them ON.
  • the "AND” gate 139 outputs A0, B0,.... N0, as shown in Figure 11B, regarding the scanning line l0.
  • the outputs are sequentially written in a second line memory 141.
  • the second line memory 141 has compartments whose number is equal to the number of horizontal windows. That is, in Figure 11B, there are N compartments of C0 through C N-1 .
  • the data A0 in the first compartment C0 of the second line memory 141 are temporarily set in the register 142, and A1 counted by the memory counter 137 is inputted through the "AND" gate 139 to the adder 140, which calculates the sum of the A0 set in the register 142 and this A1. This sum is set in the compartment C0 of the second line memory 141.
  • the sum (A0 + A1 + A2 + A3) is written in the compartment C0 of the second line memory 141
  • the sum (B0 + B1 + B2 + B3) is written in the compartment C1 of the second line memory 141
  • the sum (N0 + N1 + N2 + N3) is written in the compartment C N-1 of the second line memory 141.
  • window size in the example explained is of 16 x 4 picture elements, it is by no means limited to this example.
  • the first embodiment of the invention adjusts a chromatic color blur occurring when color image data are read both in the main scanning line and sub-scanning line directions. Hence, the color image is accurately extracted.
  • the second invention abates the effects of the color blur by averaging the chromatic color in each window comprising multiple picture elements. This ensures that the accuracy of color identification is maintained and the color form information is used as character cutout information.
  • the configuration is made so that a character and a form are distinguished by their saturation and value only. Thus, when it is not necessary to identify the color of a form, a major improvement in the processing speed is expected.

Claims (10)

  1. Appareil de traitement de données d'image en couleur (1) pour traiter des informations de teinte des éléments d'image extraits séquentiellement, ledit appareil comprenant :
    un dispositif de détection de couleur chromatique (4) pour détecter une série d'éléments d'image de couleur chromatique qui est une succession consécutive d'éléments d'image de couleur chromatique dans une ligne de balayage principale ; et
    un dispositif de remplacement (5) pour remplacer la valeur de teinte de chaque élément d'image de la série d'éléments d'image de couleur chromatique par la valeur de teinte d'un élément d'image situé approximativement sur la position au milieu de la série d'éléments d'image de couleur chromatique détectée.
  2. Appareil de traitement de données d'image en couleur selon la revendication 1, comprenant en outre :
    un dispositif de commutation de valeur moyenne de teinte (6) pour commuter la valeur de teinte remplacée de chacun des éléments d'image par la moyenne des valeurs de teinte remplacées de chaque élément d'image et des éléments d'image situés dans une zone avoisinante de chacun desdits éléments d'image.
  3. Appareil de traitement de données d'image en couleur selon la revendication 2 dans lequel ledit dispositif de détection de couleur chromatique (4) comprend :
    un premier dispositif de détection pour détecter une variation dans un élément d'image dont des données ont été appliquées séquentiellement selon la ligne de balayage principale, à partir d'une couleur achromatique en une couleur chromatique ;
    un second dispositif de détection pour détecter une variation dans un élément d'image dont les données sont appliquées séquentiellement selon la ligne de balayage principale, à partir d'une couleur chromatique en une couleur achromatique ; et
    un dispositif de compteur pour commencer un comptage d'un dispositif d'horloge pour indiquer l'élément d'image appliqué selon le résultat de détection dudit premier dispositif de détection et pour arrêter le comptage selon le résultat de détection dudit second dispositif de détection.
  4. Appareil de traitement de données d'image en couleur selon la revendication 3, dans lequel ledit dispositif de remplacement (5) comprend :
    un dispositif de réglage de valeur de teinte pour régler la valeur de teinte d'un élément d'image positionné à une position à la moitié de la valeur comptée par ledit compteur à partir de l'élément d'image sur la ligne de balayage principale à partir de laquelle ledit compteur commençait le comptage comme la valeur de teinte de ladite série d'éléments d'image en couleur chromatique ; et
    un dispositif de remplacement de valeur de teinte pour remplacer la valeur de teinte de ladite série d'éléments d'image en couleur chromatique par la valeur de teinte mise par ledit dispositif de réglage de valeur de teinte.
  5. Appareil de traitement de données d'image en couleur selon la revendication 2, dans lequel ledit dispositif de commutation de valeur moyenne de teinte (6) comprend :
    un dispositif de mémoire de ligne pour mémoriser les données de n lignes de balayage principales,
    un dispositif de comptage de moyenne pour calculer la moyenne de la valeur de teinte de n x m éléments d'image dans ladite mémoire de ligne comme une valeur de fenêtre.
  6. Appareil de traitement de données d'image en couleur selon la revendication 5, dans lequel ledit produit n x m est 5 x 15.
  7. Appareil de traitement de données d'image en couleur selon la revendication 1, comprenant en outre :
    un dispositif de conversion de signal d'image (110) pour fournir un signal d'image original en signaux de teinte et de saturation, et
    un dispositif de calcul de moyenne de teinte (112) pour calculer la moyenne de teinte dans un compartiment de fenêtre comprenant des éléments d'image multiples d'une image originale.
  8. Appareil de traitement de données d'image en couleur selon la revendication 1, comprenant en outre :
    un dispositif de conversion de signal d'image (110) pour convertir un signal d'image original en signaux de teinte et de saturation dans une unité d'éléments d'image ;
    un dispositif de calcul de moyenne de teinte pour calculer la moyenne des valeurs de teinte des éléments d'image dont la saturation est au-dessus d'un certain seuil, dans une unité d'une des multiples zones divisées ; et
    un dispositif de conversion de teinte pour convertir la teinte des éléments d'image dont la saturation est au-dessus dudit certain seuil en teinte calculée par ledit dispositif de moyenne de teinte.
  9. Appareil de traitement de données d'image en couleur selon la revendication 8, dans lequel ledit dispositif de moyenne de teinte comprend :
    un dispositif de comparaison (111) pour comparer la saturation fournie par ledit dispositif de conversion de signal d'image à une certaine valeur introduite ;
    un dispositif de circuit de porte (113) pour fournir un 0 et la valeur de teinte fournie par ledit dispositif de conversion de signal d'image, respectivement lorsque ledit dispositif de comparaison teste que la valeur de saturation introduite est plus petite ou plus grande qu'une certaine valeur ;
    un dispositif d'accumulation pour accumuler la valeur fournie par ledit dispositif de circuit de porte ;
    un dispositif de compteur (117) pour compter le nombre d'éléments d'image dont les valeurs de saturations sont plus grandes que ladite certaine valeur dudit dispositif de comparaison ; et
    un dispositif d'opération pour diviser le résultat dudit dispositif d'accumulation par la valeur de comptage dudit dispositif compteur.
  10. Appareil de traitement de données d'image en couleur selon la revendication 9, dans lequel ledit dispositif d'accumulation comprend :
    un dispositif additionneur (114) pour additionner une sortie dudit dispositif de circuit de porte appliquée à une de ses bornes d'entrée à la valeur introduite sur son autre borne d'entrée ;
    un dispositif de mémoire de ligne (115), ayant une capacité mémoire pour un balayage des éléments d'image comprenant une image, pour mémoriser la sortie dudit dispositif additionneur correspondant à l'emplacement de l'élément d'image dans un balayage ; et
    un dispositif de registre (116) pour lire et mémoriser la valeur de l'élément d'image correspondant à la sortie du dispositif de circuit de porte appliquée audit dispositif additionneur et en additionnant la valeur à l'autre borne d'entrée dudit dispositif additionneur.
EP90119238A 1989-10-09 1990-10-08 Appareil pour le traitement de données d'images en couleurs Expired - Lifetime EP0422544B1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP263676/89 1989-10-09
JP1263676A JP2746692B2 (ja) 1989-10-09 1989-10-09 色画像データ処理装置

Publications (3)

Publication Number Publication Date
EP0422544A2 EP0422544A2 (fr) 1991-04-17
EP0422544A3 EP0422544A3 (en) 1992-06-03
EP0422544B1 true EP0422544B1 (fr) 1996-05-29

Family

ID=17392798

Family Applications (1)

Application Number Title Priority Date Filing Date
EP90119238A Expired - Lifetime EP0422544B1 (fr) 1989-10-09 1990-10-08 Appareil pour le traitement de données d'images en couleurs

Country Status (4)

Country Link
US (1) US5220620A (fr)
EP (1) EP0422544B1 (fr)
JP (1) JP2746692B2 (fr)
DE (1) DE69027173T2 (fr)

Families Citing this family (30)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07101916B2 (ja) * 1991-05-14 1995-11-01 富士ゼロックス株式会社 カラー画像編集装置
US6058209A (en) * 1991-09-27 2000-05-02 E. I. Du Pont De Nemours And Company Method for resolving redundant identifications of an object
US5448652A (en) * 1991-09-27 1995-09-05 E. I. Du Pont De Nemours And Company Adaptive display system
JP3222183B2 (ja) * 1992-02-19 2001-10-22 株式会社リコー 画像処理装置
JP3302041B2 (ja) * 1992-03-05 2002-07-15 キヤノン株式会社 画像処理装置
JP3376003B2 (ja) * 1992-03-05 2003-02-10 キヤノン株式会社 画像処理装置及び画像処理方法
DE4309877C2 (de) * 1992-04-06 1994-09-15 Hell Ag Linotype Verfahren und Einrichtung zur Analyse von Farbstichen in Farbvorlagen
JP3391816B2 (ja) * 1992-08-26 2003-03-31 キヤノン株式会社 カラー画像処理装置およびその方法
US5673335A (en) * 1992-08-26 1997-09-30 Canon Kabushiki Kaisha Color image processing method and apparatus therefor
EP0593798B1 (fr) * 1992-10-19 1999-01-07 International Business Machines Corporation Méthode et dispositif pour éliminer la couleur de documents en plusieurs couleurs
JP3548205B2 (ja) 1993-10-29 2004-07-28 キヤノン株式会社 画像処理方法およびその装置
US5450217A (en) * 1994-05-23 1995-09-12 Xerox Corporation Image-dependent color saturation correction in a natural scene pictorial image
US5469276A (en) * 1994-10-11 1995-11-21 Seiko Epson Corporation Method and apparatus for reducing artifacts in halftone images using gray balance correction
US5555324A (en) * 1994-11-01 1996-09-10 Massachusetts Institute Of Technology Method and apparatus for generating a synthetic image by the fusion of signals representative of different views of the same scene
GB9426078D0 (en) * 1994-12-23 1995-02-22 Cadbury Schweppes Plc Process for manufacture of reduced fat chocolate
US6118895A (en) * 1995-03-07 2000-09-12 Minolta Co., Ltd. Image forming apparatus for distinguishing between types of color and monochromatic documents
DE19955958C2 (de) * 1999-11-19 2003-02-27 Guideguide Ag Verfahren und System zum Erstellen von Internet-Seiten und Set für ein System zum Erstellen von Internet-Seiten
US6873436B1 (en) * 2000-09-05 2005-03-29 Fuji Xerox Co., Ltd. Image processing device and recording medium
US7050086B2 (en) * 2001-06-26 2006-05-23 Pentax Corporation Electronic endoscope system with color-balance alteration process
US6944333B2 (en) * 2003-04-30 2005-09-13 Ventana Medical Systems, Inc. Color image compression via spectral decorrelation and elimination of spatial redundancy
JP4305071B2 (ja) * 2003-06-30 2009-07-29 株式会社ニコン 信号補正方法
US20090046922A1 (en) * 2005-07-04 2009-02-19 Toru Yoshikawa Surface Inspecting Apparatus
US7616343B2 (en) * 2006-04-06 2009-11-10 Kabushiki Kaisha Toshiba Image data conversion apparatus, image forming apparatus, image data conversion method and program
TWI326441B (en) * 2006-04-28 2010-06-21 Chimei Innolux Corp Driving device and driving method of liquid crystal panel
EP1898625A3 (fr) * 2006-09-06 2009-10-14 Murata Machinery Ltd. Appareil et procédé de décision de coloration d'image
JP4857173B2 (ja) * 2007-04-25 2012-01-18 日立オムロンターミナルソリューションズ株式会社 画像処理装置、画像処理方法及び画像処理プログラム
JP5349790B2 (ja) * 2007-11-16 2013-11-20 キヤノン株式会社 画像処理装置、画像処理方法、及びプログラム
US8290252B2 (en) * 2008-08-28 2012-10-16 Microsoft Corporation Image-based backgrounds for images
JP5300639B2 (ja) * 2009-07-23 2013-09-25 キヤノン株式会社 画像処理方法、画像処理装置、及びプログラム
JP6287337B2 (ja) * 2014-02-28 2018-03-07 富士ゼロックス株式会社 画像処理装置、画像処理方法、画像処理システムおよびプログラム

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4054786A (en) * 1973-09-24 1977-10-18 The United States Of America As Represented By The Secretary Of The Navy Running average computer
DE3110222C2 (de) * 1981-03-17 1985-06-20 Dr.-Ing. Rudolf Hell Gmbh, 2300 Kiel Verfahren zur partielle glättenden Retusche bei der elektronischen Farbbildreproduktion
JPS58121941A (ja) * 1982-01-13 1983-07-20 テルモ株式会社 超音波診断装置
CA1219338A (fr) * 1983-07-27 1987-03-17 Rca Corporation Circuit de traitement de signaux
US4782384A (en) * 1984-04-27 1988-11-01 Utah Scientific Advanced Development Center, Inc. Area isolation apparatus for video signal control system
US4819077A (en) * 1986-05-14 1989-04-04 Kabushiki Kaisha Toshiba Color image processing system
US4805016A (en) * 1986-08-25 1989-02-14 Kabushiki Kaisha Toshiba Endoscopic system for converting primary color images into hue, saturation and intensity images
JP2603947B2 (ja) * 1986-09-26 1997-04-23 オリンパス光学工業株式会社 原色画像間の対応領域検出装置
US4956704A (en) * 1987-10-23 1990-09-11 Canon Kabushiki Kaisha Image processing apparatus and methods for making a creative image from an original image
JPH01309467A (ja) * 1988-01-29 1989-12-13 Konica Corp 画像処理装置
JP2637498B2 (ja) * 1988-09-01 1997-08-06 富士通株式会社 画像信号処理装置
JP2641924B2 (ja) * 1988-10-31 1997-08-20 富士通株式会社 画像データ抽出装置
US5003613A (en) * 1988-12-21 1991-03-26 Recognition Equipment Incorporated Document processing system and method

Also Published As

Publication number Publication date
DE69027173D1 (de) 1996-07-04
EP0422544A2 (fr) 1991-04-17
DE69027173T2 (de) 1996-11-14
US5220620A (en) 1993-06-15
JPH03125286A (ja) 1991-05-28
EP0422544A3 (en) 1992-06-03
JP2746692B2 (ja) 1998-05-06

Similar Documents

Publication Publication Date Title
EP0422544B1 (fr) Appareil pour le traitement de données d'images en couleurs
US5134666A (en) Image separator for color image processing
US5392137A (en) Image processing apparatus in which filtering is selected for input image characteristics
EP1014699B1 (fr) Méthode pour la selection de pixels en couleur dans un bloc pour la codage en bloc tronque
US5740333A (en) Image processing method and image processing apparatus for converting an inputted color image into a two-color image
US5282026A (en) Device for judging the type of color of a document
EP0643528A2 (fr) Appareil de traitement d'image capable de réaliser un marquage d'édition
US5684600A (en) Image processor for smoothing and differentiating periodic areas and character areas using coefficients
US4996602A (en) Image-processing apparatus
EP0662765B1 (fr) Appareil et procédé de traitement d'image
EP0350930A2 (fr) Dispositif de traitement d'images
JP3473707B2 (ja) カラー画像処理装置
JPS5923969A (ja) 画信号処理装置
US6798544B1 (en) Image reading method, image reading device and image processing device
JP2851060B2 (ja) 色識別装置
JPH02204879A (ja) 色識別方法及び装置
JP2637498B2 (ja) 画像信号処理装置
JPH02137079A (ja) 色識別回路
JP2641924B2 (ja) 画像データ抽出装置
JPH06334856A (ja) 画像処理装置
JP2000341549A (ja) カラー画像処理装置
JPS60180378A (ja) 多色原稿読み取り装置
JP2638166B2 (ja) 色画像データ抽出処理装置
JP3020240B2 (ja) 色識別回路
JP3096042B2 (ja) カラー画像読取装置

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

AK Designated contracting states

Kind code of ref document: A2

Designated state(s): DE FR GB

PUAL Search report despatched

Free format text: ORIGINAL CODE: 0009013

AK Designated contracting states

Kind code of ref document: A3

Designated state(s): DE FR GB

17P Request for examination filed

Effective date: 19921120

17Q First examination report despatched

Effective date: 19940801

GRAH Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOS IGRA

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

AK Designated contracting states

Kind code of ref document: B1

Designated state(s): DE FR GB

REF Corresponds to:

Ref document number: 69027173

Country of ref document: DE

Date of ref document: 19960704

ET Fr: translation filed
PLBE No opposition filed within time limit

Free format text: ORIGINAL CODE: 0009261

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT

26N No opposition filed
REG Reference to a national code

Ref country code: GB

Ref legal event code: IF02

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: FR

Payment date: 20031003

Year of fee payment: 14

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: GB

Payment date: 20031008

Year of fee payment: 14

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: DE

Payment date: 20031016

Year of fee payment: 14

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: GB

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20041008

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: DE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20050503

GBPC Gb: european patent ceased through non-payment of renewal fee

Effective date: 20041008

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: FR

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20050630

REG Reference to a national code

Ref country code: FR

Ref legal event code: ST